Losses at every stage in the food system influence the extent to which nutritional requirements of a growing global population can be sustainably met. Inefficiencies and losses in agricultural production and consumer behaviour all play a role. This paper aims to understand better the magnitude of different losses and to provide insights into how these influence overall food system efficiency. We take a systems view from primary production of agricultural biomass through to human food requirements and consumption. Quantities and losses over ten stages are calculated and compared in terms of dry mass, wet mass, protein and energy. The comparison reveals significant differences between these measurements, and the potential for wet mass figures used in previous studies to be misleading. The results suggest that due to cumulative losses, the proportion of global agricultural dry biomass consumed as food is just 6% (9.0% for energy and 7.6% for protein), and 24.8% of harvest biomass (31.9% for energy and 27.8% for protein). The highest rates of loss are associated with livestock production, although the largest absolute losses of biomass occur prior to harvest. Losses of harvested crops were also found to be substantial, with 44.0% of crop dry matter (36.9% of energy and 50.1% of protein) lost prior to human consumption. If human over-consumption, defined as food consumption in excess of nutritional requirements, is included as an additional inefficiency, 48.4% of harvested crops were found to be lost (53.2% of energy and 42.3% of protein). Over-eating was found to be at least as large a contributor to food system losses as consumer food waste. The findings suggest that influencing consumer behaviour, e.g. to eat less animal products, or to reduce per capita consumption closer to nutrient requirements, offer substantial potential to improve food security for the rising global population in a sustainable manner.

The nexus of population growth and changing diets has increased the demands placed on agriculture to supply food for human consumption, animal feed and fuel. Rising incomes lead to dietary changes, from staple crops, towards commodities with greater land requirements, e.g. meat and dairy products. Despite yield improvements partially offsetting increases in demand, agricultural land has still been expanding, causing potential harm to ecosystems, e.g. through deforestation. We use country-level panel data (1961-2011) to allocate the land areas used to produce food for human consumption, waste and biofuels, and to attribute the food production area changes to diet, population and yields drivers. The results show that the production of animal products dominates agricultural land use and land use change over the 50-year period, accounting for 65% of land use change. The rate of extensification of animal production was found to have reduced more recently, principally due to the smaller effect of population growth. The area used for bioenergy was shown to be relatively small, but formed a substantial contribution (36%) to net agricultural expansion in the most recent period. Nevertheless, in comparison to dietary shifts in animal products, bioenergy accounted for less than a tenth of the increase in demand for agricultural land. Population expansion has been the largest driver for agricultural land use change, but dietary changes are a significant and growing driver. China was a notable exception, where dietary transitions dominate food consumption changes, due to rapidly rising incomes. This suggests that future dietary changes will become the principal driver for land use change, pointing to the potential need for demand-side measures to regulate agricultural expansion.

Model‐based global projections of future land‐use and land‐cover (LULC) change are frequently used in environmental assessments to study the impact of LULC change on environmental services and to provide decision support for policy. These projections are characterized by a high uncertainty in terms of quantity and allocation of projected changes, which can severely impact the results of environmental assessments. In this study, we identify hotspots of uncertainty, based on 43 simulations from 11 global‐scale LULC change models representing a wide range of assumptions of future biophysical and socioeconomic conditions. We attribute components of uncertainty to input data, model structure, scenario storyline and a residual term, based on a regression analysis and analysis of variance. From this diverse set of models and scenarios, we find that the uncertainty varies, depending on the region and the LULC type under consideration. Hotspots of uncertainty appear mainly at the edges of globally important biomes (e.g., boreal and tropical forests). Our results indicate that an important source of uncertainty in forest and pasture areas originates from different input data applied in the models. Cropland, in contrast, is more consistent among the starting conditions, while variation in the projections gradually increases over time due to diverse scenario assumptions and different modeling approaches. Comparisons at the grid cell level indicate that disagreement is mainly related to LULC type definitions and the individual model allocation schemes. We conclude that improving the quality and consistency of observational data utilized in the modeling process and improving the allocation mechanisms of LULC change models remain important challenges. Current LULC representation in environmental assessments might miss the uncertainty arising from the diversity of LULC change modeling approaches, and many studies ignore the uncertainty in LULC projections in assessments of LULC change impacts on climate, water resources or biodiversity.